CAST analysis of the 2003 critical baby food disastrous event in Israel
Michael Shnaid and Daniel Hartmann
Ben Gurion University of the Negev, Israel
6th STAMP Workshop 2017, MIT
2
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
The Importance of Thiamine (vitamin B1) in critical food for babies
• Thiamine plays a key role in maintaining memory, thehealth of the nervous system, and heart muscle and isimportant for growth, mental development, andlearning skills in children
• Thiamine acts as an important cofactor in metabolismand energy production.
• It is required for the biosynthesis of neurotransmittersand the production of substances used in defenseagainst oxidant stress.
3
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
The Importance of Thiamine (vitamin B1) in critical food for babies /cont.
• The only source of Thiamine in the human body isfrom food. Therefore the composition of diary-freemilk substitutes given to babies who are not breastfed,is safety critical.
• Vitamin B1 is stored in the liver in infants, but itdecreases rapidly.
• Thiamine deficiency can develop within 2–3 monthsfrom a deficient intake and can cause illness and death.
4
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
• The onset of symptoms can be very rapid and thefatality rate can be very high with death often occurringwithin a few days from the onset of symptoms.
• The symptoms of subclinical (mild or partial) Thiaminedeficiency are vague and nonspecific, making it difficultto diagnose.
• Infants who survive Thiamine-deficient related illnesseshave a poor prognosis, with motor and cognitiveimpairment and epilepsy.
The Importance of Thiamine (vitamin B1) in critical food for babies / cont.
5
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
• Infantile Thiamine deficiency (Beriberi), is rarely seentoday in developed countries after decades of strongpublic health attention.
• Beriberi poses difficult diagnostic issues and can be amissed diagnosis, as it can mimic critical illness orpolyneuropathies. In addition, clinical manifestationssuch as tachypnea, chest indrawing, tachycardia andcardiomegaly can suggest other diagnoses.
The Importance of Thiamine (vitamin B1) in critical food for babies / cont.
6
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
The Path in The USA and Globally to Perceive “Critical Food”
• Standards and regulations for infant formula and otherinfant foods aim at ensuring high levels of safety andare much more rigorous than regulations for other foodproducts.
• In fact, it was a serious safety issue that led to theadoption of the US Infant Formula Act in 1980.
• The issue was the change in the formulation of two soyinfant formula which were introduced in 1978 by amajor infant formula manufacturer in the USA.
7
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
• These deficient formulations induced metabolic illnesses andgrowth faltering in the babies who received them, and theywere later also found to be associated with adverse long-termeffects on developmental outcomes and with behavioralproblems.
• The United States Congress reviewed the matter anddetermined that protection of infants fed infant formulaeneeded to be improved by greater regulatory control overinfant formula, which led to the US Infant Formula Act of 1980.
The Path in The USA and Globally to Perceive “Critical Food” / cont.
8
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
• This act first established minimum nutrientrequirements for infant formulae, defined adulterationand quality control procedures, and specifiedinspection requirements as well as recall procedures.
• The establishment of this and further national, regional,and global standards and regulations on infantformulae and baby foods has contributed to the veryhigh level of safety of such products that exists today.
The Path in The USA and Globally to Perceive “Critical Food” / cont.
9
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
The Israeli Case-Study
• A new on-the-market vitamin B1-deficient non-dairy soy-based infant formula was marketed in Israel in 2003.
• The deficient formula, exposing infants to clinical orsubclinical B1 deficiency, was apparently sold in Israel forabout 6 months, from May 2003 to November 7, 2003.
• In November 2003, following a report of unexplainedencephalopathy in a cluster of infants in a tertiary medicalcenter in Israel, the Israeli Ministry of Health initiated aninvestigation.
10
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
The Israeli Case-Study / cont.
• The cause was found (within few days) to be a new on themarket soy-based infant formula (manufactured byHumana in Herford, Germany) for distribution in Israel byRemedia (Co Venture with Heinz), lacking vitamin B1 dueto a change in May 2003 in the formula composition.
• Thus, the Israel infantile Thiamine deficiency outbreakswas due to Thiamine deficient new soya formula, with arelative high fatality and morbidity rate and long termconsequences.
11
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
• The victims are children who were exposed to a dietbased exclusively on a vitamin B1-deficient soy-basedformula during their first year and hence have sufferedclinical or subclinical Thiamine deficiency.
• At the instant the diagnosis of beriberi was established,the deficient formula was legally withdrawn from themarket.
The Israeli Case-Study / cont.
The Marketable Dream
13
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
40
200
5000
The consequences of a lack of vitamin B1 (Thiamine) in critical food in Israel revealed over time:
• First Circle, “Immediate“ findings. 40 infants, of whom 3 died, 8 suffered very severe health impairments and the rest 29 suffered severe morbidity.
• Second Circle, after two years. About 200 babies were diagnosed with seemingly minor harm. Although these babies were given Thiamine treatment at the time, yet worsening in their development can be observed.
• Third circle, 14 years after the event, and continues. 5,000 other casualties were identified. These teenagers today suffer from development problems and various cognitive impairment ADHD.
• In total, 65% of the infants who consumed plant-based Remedia have a medical condition one way or another.
14
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Accident Definitions - Remedia disaster
• Illness and death of infants who were not nourishedby appropriate critical food, essential for theirdevelopment, health and growth.
• Illness and death of infants fed by a critical foodnegatively affecting their health.
• Illness and death of infants untreated in time whilesuffering harms due to the defective critical food.
A Thiamine deficient soy protein-based infant formula led tosevere Thiamine deficiency in recipient infants with lacticacidosis, encephalopathy, cardiomyopathy, and deaths, alongwith long-term neurologic problems in surviving children.
CA
ST
Value Chain, System’s Lifecycle
& Evidence Based Safety
Concept Requirements Design Build Operate DisposeAccident
Analysis
Solution Free Solution Oriented
16
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Proximal or Immediate Event
• In Disastrous events such as the Remediaaccident, the scope of timelines and places aremuch broader.
• The duration and location of the loss event andits consequences are not instant, and are notreflected immediately and revealed by clearfeedback loops.
• An immediate / proximal event, is defined by us as an event that representsthe very fast closing of a feedback loop in a particular type of accident.
• In reasonably "simple" accidents, for example a car , train or airplane crash, itis relatively very easy to define the tangible immediate events that precededthe accident and the moments of the accident itself.
17
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
New Loss-Events / Accidents Classification
Tangible Nontangible
Syst
emic
Eve
nt
Pro
xim
al E
ven
tKnown
Accidents
Hidden Accidents
Feedback loop
Accomplished and Known
Feedback loop Not accomplished and / or unknown
18
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Remedia Disaster
19
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Course of Phases and Events in Time
• Events were divided into five Phases "time series“.
• Critical events (in development and operation Safety Control
Systems) preceding the proximal event and spread out over a
period longer than a decade.
• Other critical events (in Response Safety Control System) that
contribute to the severity of the disaster, spread out well over
a period of more than a decade after the immediate accident.
20
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
History and "time series“ related to the Remedia DisasterPhases Time Span Duration
Activities –Hazardous Processes
The sequence of events preceding the disaster
1990 –November 2002
ca. 12 Years Normal successful Business Activities (Remedia & Humana)
The sequence of events leading to disaster
December 2002 –01.07.2003
ca. 5 Months
New Product:Design (Remedia & Humana), Production (Humana) & Marketing (Remedia)
Hidden Loss Events July – 8.11.2003 ca. 4 Months
Unsafe Consume & Health Consequences
Proximal event 08.11. – 11.11.2003 ca. 4 Days Disease Discovery & treatment onset
The sequence of events following the disaster
November 2003 –Present
ca. 14 Years Disease Progress
21
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
SDCS
Legislation
Regulation
Company Management
Hazardous Process
SOCS
Legislation
Regulation
Company Management
SRCS
Legislation
Regulation
Management
Hazardous Process
Hazardous Process
Project Management
Project Management
Project Management
Legend:Control ChannelFeedback ChannelInformation Channel
Legend:Control ChannelFeedback ChannelInformation Channel
The Complete Sociotechnical System as hierarchical control structures
22
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Hazards (SDCS, SOCS & SRCS):
• H1 – infant critical safety formulas and products are designed and produced in an unsafe way
• H2 – Infants are fed with unsafe critical food
• H3 – Critical baby food is given in an unsafe way
• H4 – Infants do not receive safe critical food for their healthy development
• H5 – Infants do not receive adequate treatment to recover from damage caused by lack of critical food
23
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Safety Constraints:
• SC5 - When babies are fed non-safe critical food, appropriate medical measures and protocols must be take
• SC1a - Infant formulas and products must be designed in a way that willgive infants who consume them good health along their growthperiod
• SC1b - Infant formulas and products must be produced in a way that will giveinfants who consume them good health along their growth period
• SC2 - Food products for infants which are on the shelf must be safe and effective
• SC3 - Babies must be fed infant nutrition products in a beneficial way for their health and their normal development
• SC4 - Babies must be fed infant nutrition products in a safe and effective manner
24
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Hazards Safety Constraints Violation
H1 – Infants are fed with unsafe critical food
SC1 - infant formulas and products must be designed and produced in a way that babies develop healthy along their growth periodSC2 - infants food products available on the market must be safe and effectiveSC3 - Babies must be fed infant nutrition products in a beneficial way for their health and their normal developmentSC4 - Babies must be fed infant nutrition products in a safe and effective manner
SC1a - SDCS
SC1b - SDCS
SC2 - SOCS
SC3 - SOCS
SC4 – SOCS
H2 - Critical baby food is given in an unsafe way
SC3 - Babies must be fed infant nutrition products in a beneficial way for their health and their normal developmentSC4 - Babies must be fed infant nutrition products in a safe and effective manner
SC3 - SOCS
SC4 - SOCS
H3 – Infants do not receive safe critical food for their healthy development
SC3 - Babies must be fed infant nutrition products in a beneficial way for their health and their normal developmentSC4 - Babies must be fed infant nutrition products in a safe and effective manner
SC3 - SOCS
SC4 – SOCS
H4 – Infants do not receive adequate treatment to recover from damage caused by lack of critical food
SC5 - Appropriate medical protocols and measures must be taken to ensure the remediation of diagnosed babies fed with non-safe critical food.
SC5 - SRCS
25
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Development Control Structure
System Development - Germany
European Parliament
German Parliament
German Government
Agencies – Associations - Insurance
WHO
Humana
Specialists
Quality Management
Production purchase Design
Legislation
Regulation
Management
Hazardous Processes
Project Management
Legend:Control ChannelFeedback ChannelInformation Channel
System Development - Israel
Israeli Parliament
Israeli Government
Agencies – Associations - Insurance
Health Treasury Economy
Remedia
Specialists
Quality Management
Design Process
Committees
26
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Operation Control Structure
Israeli Parliament
Israeli Government
Regulatory Agencies, Health Associations, Insurances
Remedia
Distribution companies
Consumers - Babies
Consumers - Parents
Commitees
Legislation
Regulation
Management
Project Management
Hazardous Process
27
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Respond Control Structure
Israeli Parliament
Israeli Government
Regulatory Agencies ,Medical Institutions , Insurances
Health Organizations
Babies Physiology
Sick babies
Committees
HospitalsHealth
ServicesClinics
Health Insurances
Legislation
Regulation
System Management
Project Management
Hazardous Process
28
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Development (Design Phase) Control Structure
Physical Level
Design Process Humana
Formula & Algorithm
ActuatorDesign Quality Management
29
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Development (Production Phase) Control Structure
Physical Level
Production Humana
Formula Production
ActuatorProduct Quality
Management
External Materials
External Sevices
30
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Operation Control Structure
Physical Level
Consumers DistributersParents
Helpless babies
RemediaShops
Legend:Control ChannelFeedback ChannelInformation Channel
31
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
System Response Control Structure
Physical Level
Consumers Medical sources
Parents
Helpless babies
ClinicsHospitals
Baby CentersPhysicians
Legend:Control ChannelFeedback ChannelInformation Channel
Legend:Control ChannelFeedback ChannelInformation Channel
32
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Control input or external information
wrong or missingController
Actuator Sensor
Controller 2Controlled Process
Process Model inconsistent,
incomplete, or incorrect
Inadequate Control Algorithm(Flaws in creation, process
changes, incorrect modification or adaptation)Inappropriate,
ineffective or missing control action
Inadequate operation
Delayed operation
Conflicting control action
Process input missing or wrong Unidentified or out of
range disturbance
Process output contributes to system hazard
Incorrect or no information provided
Measurement inaccuracies
Feedback delays
Inadequate operation
Inadequate or missing feedback
Feedback delays
Weak Control Structure
Component failuresChanges over time
33
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
General Conclusions as to SDCS Physical Level
• Very complex and dispersed ST control structures.• The organizations involved do not understand the full
range of hazards and their potential consequences.• No clearly defined Accountability and Responsibility in
the SDCS.• Structural weakness causing inconsistency in collecting
and processing critical data, and in real time.• Synchronization and Coordination problems due to
structural problems.
34
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
General Conclusions as to SDCS Upper Levels
• Inefficient, bureaucratic ST systems.• Regulations are outdated, inefficient and not
involved in real life.• A chronic shortage of resources to carry out
policy.• No effective inspection and enforcement.• No change and / or drift management, and no
understanding of their meaning regarding safetyimpact on the systems.
35
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
General Conclusions from SOCS Physical Level
• The organizations and persons involved do notunderstand the full range of hazards and theirpotential consequences.
• No clearly defined Accountability and Responsibilityin the SOCS.
• Structural weakness causing inconsistency incollecting and processing data in real time.
• Synchronization and Coordination problems due tostructural problems.
36
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
General Conclusions from SOCS Upper Levels
• Inefficient, bureaucratic systems.• Regulations are outdated, inefficient and not
involved in in real life.• A chronic shortage of resources to carry out policy.• No effective inspection and enforcement.• No change and / or drift management, and no
understanding of their meaning regarding safetyimpact on the systems.
37
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
Lessons Learned
• The need for Evidence Based Safety:
o To achieve a “safer world”, input from accident analysis is crucialfor any system hazard analysis.
• The entire Sociotechnical System comprising of SDCS, SOCS and
SRCS and all their interactions should be considered and analyzed.
• Apart of the Physical level, most upper levels in the Socio-Technical
System can contribute generic Hazards, Safety Constrains andproblems that can be defined and contribute to any Hazard analysisof similar systems and domains of operation.
• The importance of Hazard analysis for any crucial change in systems.
38
Introduction | Case Study | Definitions | Timeline | Control Structures | Results | Conclusions ||| Shnaid & Hartmann 2017
?
?
?
Thank You
Daniel [email protected]
Michael Shnaid
Top Related